Drum Bus Compression

J

JohnnyRecs

New member
I am wondering specifically about release times for drum bus compression. In what instances is a fast release time preferred, vs. a slow release time, when using a drum bus compressor to glue the elements together, and get things pumping together tight?

I've been leaving my kick out of my drum bus since compressing it usually kills a lot of low end. So I send the kick straight to the master. I got this idea from Steve Duda in his reddit AMA. So drum bus is snare, hats, percussion, etc.

I am thinking that I'd like 2 compressors on the drum bus. One responding to the program material (snare, hats, percussion), to glue it together and keep things tightly grooving (around 3db of GR); and a second one that's side chained to the kick to get the entire "kit" pumping/making room for the kick (also around 3db of GR). Thoughts on this approach?

I watched an Instagram video of Deadmau5 using an API 2500 on his drum bus and it looks as if the release time is set to 500ms (3db of GR), and it's pumping with the kick, which makes me think the compression is triggered by kick from a side chain (assuming his kick is not also being sent to the drum bus, like Steve Duda recommended)? Any thoughts on his use of compression here? Instagram

Thanks, and very interested in the approach others take to drum bus compression, specifically in terms of release times, pumping, side chaining, etc.
 
Interesting. I will subscribe to this thread. I'm not confident enough in my skills to contribute all that much to this thread at this point. But I will say though that effective way to spice up your drum bus is through parallel processing. It can really do a lot for your drums.

I'd have to say though that the release time depends entirely on what is actually triggering the compression, how many different sounds do you have in your drum bus that exceeds the compressor threshold, and if you should let the highest peak and the release time on it to dictate what happens to the rest of your drum bus.

Id say that it is highly on a track to track basis, but I would say that a short release time would be better if you don't want the highest peak in your drum bus to squeeze too much out of the rest, and the actual time of the release, dictates how fast or slow the rest will pump with it. There is no right way to do it.

Just realized I had to say plenty in this post.. haha
 
Last edited:
Like its been said up there, you aren't gonna figure out anymore from this thread regarding your release time then you would by just playing with your release knob for a few seconds..play with the release knob for more than a few seconds and youll have figured out more than this thread can tell you about it. Its all gonna change track by track. You won't find one release time thats gonna work for everything. As far as keeping the kick out, go for it, I've seen it a lot. Starting to do it like that too.

And your side chain cant go on the drum bus, it has to go on whatever you are side chaining to your kick. With that being said I assumed you'd be side chaining a pad or something to your kick. BUT if you are side chaining the other drums to your kick, then yes it would go on the bus too..but if thats the case you can just use the same compressor to compress and side chain the kit, you don't need 2..Just turn side chain on on your compressor and set the kick to the key input. If you have problems programming the side chains perimeters without changing the original compressors settings you could use two maybe, but I don't think you wanna be side chaining all your drums THAT much anyway.
 
Last edited:
Yumid said:
your side chain cant go on the drum bus, it has to go on whatever you are side chaining to your kick. BUT if you are side chaining the other drums to your kick, then yes it would go on the bus too..
Click to expand...

Please clarify this..

You can easily use 2 compressors. The sidechained one reacts to an incoming signal outside the bus, and the other one compresses the sidechained output. It would produce dramatically different results than to try and shape your drum bus using only one sidechained compressor.
 
Wallengard said:
Yumid said:
And your side chain cant go on the drum bus, it has to go on whatever you are side chaining to your kick. With that being said I assumed you'd be side chaining a pad or something to your kick. BUT if you are side chaining the other drums to your kick, then yes it would go on the bus too..
Click to expand...
Please clarify this..

You can easily use 2 compressors. The sidechained one reacts to an incoming signal outside the bus, and the other one compresses the sidechained output. It would produce dramatically different results than to try and shape your drum bus using only one sidechained compressor.
Click to expand...

so I read this and thought wtf - you would side-chain any drums to the kick on their individual channels not on the group buss - if you try to side-chain the drum buss to the kick (which includes the kick) you would end up dropping the kick which would increase the need for side-chaining which would drop the kick which would increase the need for side-chaining and so on - this would be a positive feedback loop that can end up running away on you
 
bandcoach said:
so I read this and thought wtf - you would side-chain any drums to the kick on their individual channels not on the group buss - if you try to side-chain the drum buss to the kick (which includes the kick) you would end up dropping the kick which would increase the need for side-chaining which would drop the kick which would increase the need for side-chaining and so on - this would be a positive feedback loop that can end up running away on you
Click to expand...

Haha priceless. We managed to baffle the almighty bandcoach. ;-)

The scenario in my head went like this;

A drum bus, with a Compressor on it, sidechained to a kick that is not part of the drum bus. Then a Compressor in the chain after the sidechained compressor to compress the signal coming out of the sidechained one.

Not sure whether you got that already but, that's how I meant to say it
 
not baffled just amazed that anyone would try to side-chain the drum buss to the kick, which by definition must be on the drum buss
 
JohnnyRecs said:
I am wondering specifically about release times for drum bus compression. In what instances is a fast release time preferred, vs. a slow release time
Click to expand...

That would depend completely on the sounds being compressed and the tempo of the track.



JohnnyRecs said:
I've been leaving my kick out of my drum bus since compressing it usually kills a lot of low end. So I send the kick straight to the master. I got this idea from Steve Duda in his reddit AMA. So drum bus is snare, hats, percussion, etc.

I am thinking that I'd like 2 compressors on the drum bus. One responding to the program material (snare, hats, percussion), to glue it together and keep things tightly grooving (around 3db of GR); and a second one that's side chained to the kick to get the entire "kit" pumping/making room for the kick (also around 3db of GR). Thoughts on this approach?
Click to expand...

do it and see how you like it.



JohnnyRecs said:
I watched an Instagram video of Deadmau5 using an API 2500 on his drum bus and it looks as if the release time is set to 500ms (3db of GR), and it's pumping with the kick, which makes me think the compression is triggered by kick from a side chain (assuming his kick is not also being sent to the drum bus, like Steve Duda recommended)? Any thoughts on his use of compression here? Instagram
Click to expand...



The compressor is clearly NOT being triggered by the kick alone in this situation, as the hat can clearly be seen affecting the gain reduction on the meters.


Also, the release time setting is the thing that will give you the least amount of information to help you as it is most instrument and tempo dependent.

More important things with which to concern yourself?
-threshold (the number is not important, but rather WHERE IT IS HITTING)
-ratio
-knee
-attack (though also dependent on the envelope of the instrument)
-filter
-"old/new" (specific to API in general)
-"detector thrust" (specific to API2500)
 
Wallengard said:
The scenario in my head went like this;

A drum bus, with a Compressor on it, sidechained to a kick that is not part of the drum bus. Then a Compressor in the chain after the sidechained compressor to compress the signal coming out of the sidechained one.
Click to expand...

I believe he is describing the opposite arrangement...

Compress the "drums-minus-kick"...

Then put a second compressor on the "drums-minus-kick" bus which is being sidechain triggered by the kick (which is outside of the bus)...

This second compressor will take the result of the first compressor and "duck" it when the kick hits, thereby "making room for the kick".
 
The release of the compressor acts differently depending on which compressor you're using, e.g. an SSL-type bus compressor will give you something different to a 670-type or an API 2500. What I personally do if I don't know which I prefer is to load up the two I'm trying to decide between and set them to they go a little heavy on the GR (4-8 dB or more), try a few different release settings and shift between the two to see how they react with the material. The sound of a quick or slow release might also vary depending on the material (esp. with the vari-mu type comps), and especially on a drum bus, so to try it out really seems like the best advice.

It often seems they end up having a side-effect on the frequency spectrum as well, e.g. the API 2500 might seem to push forward the mids a little while the 670 does something to the low end.

As for excluding your kick from the drum bus and ducking the drum bus with the kick, that's definitely something you can do. But making decisions on how it sounds, you should make it in the context of the track in order to tell how the compression reacts with all the other elements in the track, as opposed to sitting with a looped drumloop over and over again. :P
 
You must log in or register to reply here.
Back
Top